Huntington's disease (HD) is one of several dominant neurodegenerative diseases caused by a similar toxic gain of function mutation in the disease protein: expansion of a polyglutamine (polyQ)-encoding tract. Currently, no therapy exists for HD. RNA interference (RNAi) has emerged as a leading method to reduce disease gene expression by targeting and degrading the encoding mRNA. Our preliminary work demonstrates that vector-mediated RNAi can reduce huntington (Htt) expression and improve disease phenotypes in mouse models of disease. We used minimal off-target silencing as a primary objective in vector design, taking advantage of bioinformatics and microarrays to identify transcriptional consequences of the inhibitory RNAs. These vectors, though minimized for off-sequence silencing, still reduce expression of both HTT alleles. Recent analysis of the HD population suggests that there are 4-5 predominant single nucleotide polymorphisms (SNPs) representing almost 90% of HD patients. This data gives us the opportunity to test if vectors for allele-directed silencing with minimized off-targeting are safe and effective in vivo. For this, we developed novel transgenic mice, with the mutant transgenes engineered to contain these relevant SNPs. We propose to now test if artificial miRNAs targeting disease-linked polymorphisms can preferentially silence mutant htt alleles in vivo. Finally, we have used data on transcriptional dysregulation in HD to identify, build and preliminarily test endogenously regulated promoters for expression control of inhibitory RNAs to further improve their safety profile. We will now test the capabilities of these promoters to drive RNAi in a disease responsive manner in vitro and in vivo.